14.2 The testbed systems

 < Free Open Study > 



14.2 The testbed systems

To ensure that all the databases were tested on a level playing field, the four PCs used were configured with identical hardware. The hardware for each of the machines is defined in Table 14.1.

Table 14.1: Testbed Configuration

CPU:

Pentium III @ 500 MHz

Total RAM:

256 MB

Operating System:

Windows NT 4.0 Service Pack 5

To ensure that all of the machines were equal in the sense of performance capabilities, an independent performance assessment group ran a PC performance benchmark to validate each machine's performance specifications before the actual database benchmark tests were performed. The results are discussed in the next section.

14.2.1 PC performance assessment benchmark

The PC computer architecture performance test utilized is comprised of 22 individual benchmark tests that are available in six test suites. The six different test suites test for the following:

  • Integer and floating-point mathematical operations

  • Tests of standard two-dimensional graphical functions

  • Reading, writing, and seeking within disk files

  • Memory allocation and access

  • Tests of the MMX (multimedia extensions) in newer CPUs

  • A test of the DirectX 3D graphics system

The test results reported are shown as relative values. The larger the number the faster the computer. For example, a computer with a result of 40 can process roughly twice as much data as a computer with a result of 20. The Passmark rating is a weighted average of all the other test results and gives a single overall indication of the computer's performance. The bigger the number the faster the computer. The results we observed are shown in Table 14.2.

Table 14.2: Testbed Architecture Performance Results

Parameter Tested

Oracle System

Informix System

SQL System

DB2 System

Math—Addition

96.6

96.2

94.6

97.0

Math—Subtraction

96.4

97.1

96.2

97.6

Math—Multiplication

101.1

101.4

101.4

103.1

Math—Division

12.9

12.8

12.9

13.0

Math—Floating-Point Addition

87.7

87.8

87.6

88.7

Math—Floating-Point Subtraction

89.4

89.5

88.6

90.1

Math—Floating-Point Multiplication

91.7

91.7

90.9

92.3

Math—Floating-Point Division

14.8

14.8

14.8

14.9

Math—Maximum Mega FLOPS

171.2

172.2

170.7

177.6

Graphics 2D—Lines

17.5

17.6

17.5

17.8

Graphics 2D—Bitmaps

12.9

12.9

12.8

12.9

Graphics 2D—Shapes

4.7

4.7

4.7

4.7

Graphics 3D—Many Worlds

22.9

23.0

22.9

22.9

Memory—Allocated Small Blocks

86.6

87.6

87.0

87.6

Memory—Read Cached

67.9

68.4

68.0

68.5

Memory—Read Uncached

48.7

48.8

50.0

49.1

Memory—Write

40.8

41.1

40.9

41.4

Disk—Sequential Read

3.2

3.8

3.7

3.1

Disk—Sequential Write

2.9

3.4

3.4

2.9

Disk—Random Seek

1.2

2.3

3.6

2.1

MMX—Addition

97.7

94.5

97.8

99.4

MMX—Subtraction

92.3

98.2

93.3

96.0

MMX—Multiplication

97.8

97.5

96.9

99.1

Math Mark

75.6

75.8

75.2

76.8

2D Mark

46.7

46.9

46.7

47.1

Memory Mark

58.7

59.2

59.2

59.4

Disk Mark

19.3

25.1

28.4

21.5

3D Graphics Mark

15.5

15.7

15.5

15.6

MMX Mark

48.8

49.2

48.9

50.0

Passmark Rating

45.7

47.2

47.8

46.7

Assessment of results

The performance assessment test found that the computer system configured for the DB2 servers appeared to have better performance than the other systems in most of the tests. However, the Passmark rating (weighted average of all test results giving a single overall indication of performance) of the computer system configured for the SQL Server 2000 was the highest.

14.2.2 Burn-in test

The computer system hardware burn-in test is a thorough exercise of the hardware in a PC performed over a short period of time, in the same way as normal applications use a PC over a long period of time. The burn-in test assesses the following items:

  • CPU

  • Hard drives

  • CD-ROMs

  • Sound cards

  • 2D graphics

  • 3D graphics

  • RAM

  • Network connections and printers

Burn-in test assessment

From the results shown in Table 14.3, we can see that the SQL Server 2000's CPU speed is 0.1 MHz less than the other machines. This difference is not significant, unless the duration of performance assessment tests spans a long period of time. All other measurements indicated the four testbed machines were equivalent.

Table 14.3: Burn-In Test Results

System Information:

Informix

Oracle

DB2

SQL Server

Operating System:

Win NT4

Win NT4

Win NT4

Win NT4

Number of CPUs:

1

1

1

1

CPU Manufacturer:

Intel

Intel

Intel

Intel

CPU Type:

Celeron

Celeron

Celeron

Celeron

CPU Features:

MMX

MMX

MMX

MMX

CPU Serial #:

N/A or disabled

N/A or disabled

N/A or disabled

N/A or disabled

CPU1 Speed:

501.3MHz

501.3MHz

501.3MHz

501.2MHz

CPU Level 2 Cache:

128 KB

128 KB

128 KB

128 KB

RAM:

267,821,056 Bytes (256 MB)

267,821,056 Bytes (256 MB)

267,821,056 Bytes (256 MB)

267,821,056 Bytes (256 MB)

Color Depth:

24

24

24

24



 < Free Open Study > 



Computer Systems Performance Evaluation and Prediction
Computer Systems Performance Evaluation and Prediction
ISBN: 1555582605
EAN: 2147483647
Year: 2002
Pages: 136

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net